A team of researchers from the at H won first place in the world champions this weekend.
The victory marked the first time a Canadian team has ever won the competition, which is the largest artificial intelligence (AI) soccer simulation in the world.
RoboCup uses soccer simulation to promote robotics and AI research with the research findings used to advance many real-world areas. By 2050, the competition aims to train a team of fully autonomous humanoid robots to win a soccer game against the winner of the most recent World Cup.
Team CYRUS was led to victory against reigning world champions Japan-based HELIOS2021 by Nadar Zare and Mahtab Sarvmaili, a research assistant and PhD student in the Faculty of Computer Science. Their methods use a variety of AI and machine-learning models to train autonomous ‘players’ in different positions to collaborate and play together in an intelligent way, often mimicking the behaviour of the world’s best soccer teams and players.
“This is an incredible achievement for Team CYRUS and for H,” says Stan Matwin, Canada Research Chair and director of the Institute for Big Data Analytics. “It demonstrates that Nova Scotia is really starting to lead the way in areas of AI and machine-learning research. This is an emerging way of using AI where we use a whole range of methods to create a simulation of perception so that the agents or ‘players’ work autonomously but also learn how to collaborate and could be applied to endless fields and situations.”
While these methods are presenting new ways to play and consume sports particularly during a time where much in person activity has been suspended, the team is using their research to help humans improve their soccer techniques through an emerging project with Halifax Wanderers and Dal Varsity. This collaboration will help the teams better understand their players and team dynamics through data analysis using some of the AI methods developed for RoboCup.
The final RoboCup 2021 round can be viewed below and on the competition’s .